1 Industry, firm, year, and country effects on profitability in EU food processing This paper decomposes the variance in food industry return-on-assets into year, country, industry, and firm effects. Besides these main effects, we include several interactions and discuss their theoretical foundations. After determining effect significance in a nested ANOVA with a rotating pattern of effect introduction, we estimate effect magnitude using components of variance in a large sample of corporations. The results show that firm characteristics are more important than industry structure in determining food industry profitability in Europe. Main effects and interactions of year and country membership are weak, indicating that performance differentials can poorly be explained by macroeconomic and trade theory.
Key words: ROA, decomposition, variance components, MBV, RBV.Running head: Industry, firm, year, and country effects on profitability JEL: L00, C22
Introduction
'There are many theories because each is based on different assumptions about the world; it is theirrelevance rather than their logic which is in dispute. ' (Cook, 1958: 16).In a perfectly competitive market, firm performance that deviates from the average should not exist in the long run. However, such deviations are not an exception to the rule but in fact the norm, especially in industries characterized by high sunk costs or other impediments to competition as the food sector seems to be. The ability of firms to 2 earn returns persistently above the norm has been widely analyzed.1 While the so-called 'market-based view', which draws heavily on Industrial Organization (IO) theory, mainly attributes such 'abnormal' profits to industry characteristics, proponents of the 'resource-based view' assume that performance differentials can be better explained by firm properties. 2 In order to resolve this debate, a series of contributions following Schmalensee's (1985) seminal paper has used components-of variance analysis (COV) and nested (i.e. hierarchical) analysis of variance (ANOVA) techniques to decompose the variation in firm profitability into firm and industry specific effects. Subsequent papers have also looked at the impact of year and, more recently, of country effects on firm profitability. While the influence of country and country-industry interactions on the variation in profitability can be explained by models developed in trade theory, the aforementioned body of literature has paid little attention to the theoretical foundations 1 (e.g. Barney 1991, Bowman and Helfat 2001, Brush, Bromiley and Hendrickx 1991, Geroski and Jaquemin 1988, Gschwandtner 2005, Goddard and Wilson 1999, Mahoney 1995, McGahan and Porter 1997, 1999, 2003, Mueller 1977, 1990, Odagiri and Maruyama 2002, Roquebert, Phillips and Westfall 1996, Rumelt 1991, Schmalensee 1985, Teece, Pesano and Shen 1997, Waring 1996, Werenfelt 1984.2 Examples for studies that support the 'market-based view' are: Caves and Porter (1977), McGahan and Porter (1999), Schmalensee (1985), Slat...